With the explosion of data in recent years, storage technologies continue to be a critical infrastructure in the world. It is predicted that 44 zettabytes (44 billion terabytes) will be generated in 2020. While most data will be stored on hard disks, significant data will be stored on solid-state disks (SSD). SSDs contain flash-based memory.

The book provides an in-depth look at flash memory, starting with an introduction that includes the physical layout of a chip. Basic operations like read, write, and erase are explained, with page size, block size, latency, and power constraints discussed via tables. Techniques for garbage collection and wear leveling specific to flash memory are dealt with in detail. Many simulation software programs and their use in developing micro and macro benchmarks are discussed in detail, encouraging readers to develop on their own as well.

There are several chapters on flash file systems. Such systems are compared using various performance metrics; several charts and graphs are included. This research can be used to improve file system code in popular open-source operating systems like Linux and FreeBSD. Improving flash performance has a direct effect on end user response time because flash is used to cache at the last step. Many servers are even moving to all-flash storage with the falling flash prices.

This well-researched book is very good for industry practitioners and research students. While it is not a traditional textbook--it lacks exercises--it does include a good list of references.